Genome-Wide Epigenetic Regulation of Mirnas in Cancer

Published OnlineFirst January 10, 2013; DOI: 10.1158/0008-5472.CAN-12-3731

Cancer Review Research

Genome-Wide Epigenetic Regulation of miRNAs in Cancer

Constance Baer1, Rainer Claus2, and Christoph Plass1

Abstract Aberrant microRNA (miRNA) expression contributes to tumorigenesis and cancer progression. Although the number of reported deregulated miRNAs in various cancer types is growing fast, the underlying mechanisms of aberrant miRNA regulation are still poorly studied. Epigenetic alterations including aberrant DNA methylation deregulate miRNA expression, which was first shown by reexpression of miRNAs upon pharmacologic DNA demethylation. However, studying the influence of DNA methylation on miRNA transcription on a genome-wide level was hampered by poor miRNA promoter annotation. Putative miRNA promoters were identified on a genome-wide level by using common promoter surrogate markers (e.g., histone modifications) and were later validated as such in different tumor entities. Integrating promoter datasets and global DNA methylation analysis revealed an extensive influence of DNA hyper- as well as hypomethylation on miRNA regulation. In this review, we summarize the current knowledge of the field and discuss recent efforts to map miRNA promoter sequences and to determine the contribution of epigenetic mechanisms to the regulation of miRNA expression. We discuss examples of tumor suppressive and oncogenic miRNAs such as the miR-34 and miR-21 family, respectively, which highlight the complexity and consequences of epigenetic miRNA deregulation. Cancer Res; 73(2); 1–5. Ó2012 AACR.

Introduction and DNMT3B). Methylation of promoter-associated CpG dinu- MicroRNAs (miRNA) are small noncoding RNAs that regu- cleotides (especially in CpG islands) usually correlates with late gene expression posttranscriptionally by binding to com- reduced transcription levels of the respective gene; promoter plementary sequences in the 30 untranslated region (UTR) of hypermethylation of tumor suppressor genes causes their messenger RNAs (mRNA). The interaction results in mRNA silencing in cancer cells (2). In tumors, miRNAs were suggested cleavage or inhibited protein synthesis, leading to reduced to be deregulated by similar mechanisms, yet poor annotation protein expression of the targeted gene. Thus, up- or down- of miRNA promoters hampered global analyses of their regu- regulation of miRNAs influences the expression of oncogenes lation (3). The miRNAs are either located outside of other and tumor suppressor genes. The miRNA expression profiling transcripts and thus are transcribed independently (intergenic experiments were carried out in a number of different cancer miRNAs) or reside within the intron of a host gene (intragenic) fi entities and identified a large set of aberrantly regulated and can be cotranscribed. However, miRNA-speci c promo- miRNAs, of which many were deregulated in several tumor ters have also been described for intragenic miRNA (4). entities. The miR-21, for example, targets the tumor suppressor Maturation of miRNAs involves several processing steps gene PTEN in lung cancer and other malignancies, and miR- starting from a primary transcript, which may vary in length 15a/16-1 repress the antiapoptotic BCL2 in prostate carcinoma from several 100 base pairs (bp) to far more than 10 kb. Primary and chronic lymphocytic leukemia (CLL; ref. 1). The molecular transcripts may contain several polycistronically arranged mechanisms regulating normal miRNA expression or causing miRNAs (5) and introns and exons of protein-coding genes. their deregulation in malignant diseases are mainly unknown. For example, the 33.8 kb transcript including miR-34a contains Epigenetic modifications, such as aberrant CpG methyla- one 30 kb intron, and only a small part of the second exon tion, contribute to deregulated gene expression in cancer cells. encodes the mature miR-34a (6). Most primary miRNAs are – Methyl groups are transferred to the carbon atom at position 5 cleaved in the nucleus by the RNase complex Drosha DGCR8 of cytosines by DNA methyltransferases (DNMT1 or DNMT3A to an approximately 70 nt precursor miRNA (pre-mir). Pre-mirs are exported to the cytosol, where they are processed by the endoribonuclease Dicer giving rise to the mature miRNA (21– Authors' Affiliations: 1Department of Epigenomics and Cancer Risk 22 nt), which ultimately regulates protein expression (1). Factors, German Cancer Research Center (DKFZ), Heidelberg; and 2Department of Hematology/Oncology, University of Freiburg Medical Defects in miRNA processing can contribute to aberrant Center, Freiburg, Germany expression of mature miRNAs in cancer. For example, DICER1 Corresponding Author: Christoph Plass, German Cancer Research Cen- mutations were observed in pleuropulmonary blastoma (7) or ter (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. Phone: tumors of the reproductive system including nonepithelial 49-6221-42-3300; Fax: 49-6221-42-3359; E-mail: [email protected] ovarian cancer and embryonal rhabdomyosarcomas (8). doi: 10.1158/0008-5472.CAN-12-3731 TARBP2 is part of the Dicer-containing complex and carries Ó2012 American Association for Cancer Research. mutations in hereditary carcinomas with microsatellite

www.aacrjournals.org OF1

Baer et al.

instability (9). In the same tumor entities, miRNA maturation miRNAs that were identified as targets of epigenetic silencing can be impaired by mutations in EXPORTIN-5, which mediates is reviewed in refs. 13 and 18. the nuclear export of precursor miRNAs (10). A common characteristic of most candidate-based studies is In the cell, primary miRNas have a very short half-life due the focus on miRNAs that are closely associated with CpG to their rapid processing and, hence, are rather difficult to islands. However, similar to protein-coding genes, DNA meth- detect. However, to study epigenetic regulation of miRNAs, ylation–dependent regulation of miRNAs does not necessarily knowledge of the transcriptional start sites (TSS) of primary require the vicinity of CpG islands. An example is miRNA-199a. transcripts is required. Upstream of miR-199a-2, located in the intron of DNM3, a 390 bp stretch, not a CpG island, was differentially methylated in a testicular cancer cell line. The DNA methylation level in the miRNAs are Deregulated by Aberrant DNA identified region inversely correlated with expression of miR- Methylation 199a-5p and with tumor malignancy (19). In cell lines, pharmacologic inhibition of DNMTs causes To comprehensively characterize epigenetic regulation of DNA demethylation, which leads to reactivated expression of miRNAs genome-wide and independent of CpG islands, a some miRNAs, suggesting DNA methylation as the underlying systematic approach for large-scale identification of primary cause of repression. As an example, treatment of the human miRNA transcripts and their promoters was required. bladder cancer cell line T24 with the DNMT inhibitor 5-aza-20- deoxycytidine led to global demethylation. In combination fi with inhibition of histone deacetylases by 4-phenylbutyric Identi cation of miRNA Promoters acid, expression of 17 miRNAs could be restored. Among The first primary miRNA transcripts were experimentally them, miR-127 displayed a prominent transcriptional upre- identified by northern blot or rapid amplification of cDNA- gulation, and was shown to target the proto-oncogene BCL6 ends (RACE) leading to the discovery of transcripts that are (11). A similar approach identified epigenetically (de)regulat- longer than the precursor miRNA and cover its sequence. For ed miRNAs by profiling miRNA expression in colorectal example, a 600 nt RNA molecule constituting the primary cancer (CRC) cells genetically deficient for DNA methyltrans- transcript of miR-30a was identified by northern blot using ferase (DNMT)1- and DNMT3B. In this study, miR-124 was a set of probes surrounding miR-30a (5). identified as epigenetically silenced in CRC (12). Follow-up However, hybridization- or amplification-based approaches studies could show that miR-124 is epigenetically silenced in are restricted to individual, short, and highly abundant primary at least 10 different tumor entities including hematopoietic transcripts. To overcome these limitations, surrogate markers malignancies as well as gastric, liver, cervical, and breast for promoter sequences were used in "miRome"-wide screens cancer (13). (4, 20, 21). Polymerase II (Pol II) transcribes not only protein Because of the limited annotation of miRNA promoters, coding genes but also a majority of known primary miRNA miRNA located in CpG islands or in close vicinity were among transcripts (22). Thus, detection of Pol II-binding regions the first candidates to study epigenetic regulation. MiR-9-1 is allows identification of DNA stretches as miRNA promoters associated with a CpG island 200 bp upstream, which is (4). Another surrogate marker for promoter presence (includ- hypermethylated in breast cancer as well as in cell lines ing that of miRNA promoters) is trimethylation of histone 3 at derived from lymph node metastasis of CRC, melanoma, and lysine 4 (H3K4me3; refs. 4, 20, 21). Both Pol II- and H3K4me3- head and neck cancer (reviewed in ref. 13). CpG islands are also associated genomic regions are detectable by chromatin located adjacent to the miR-200 family members, which are immunoprecipitation (ChIP). Recent ChIP approaches allowed master mediators of the epithelial phenotype. CpG island comprehensive detection of such regions on a genomic scale methylation correlated with downregulated miRNA expres- (4, 20). Integration of supporting data such as additional sion in breast and prostate cancer cell lines (14). In addition, promoter features into bioinformatical models increases the loss of miRNA expression was linked to acquisition of mesen- confidence for miRNA promoter detection. These additional chymal features as observed in epithelial–mesenchymal tran- data include information on the overlap of putative promoters sition and tumor progression (15). Another early discovered with start sites of expressed sequence tags, sequence conser- example of silenced miRNAs is miR-1-1. The suspected primary vation (20), or the annotation of cap analysis gene expression transcript of miR-1-1 is C20orf166; the CpG island covering (CAGE) tags (21). Recently, the Encyclopedia of DNA Elements exon 1 and intron 1 of C20orf166 was found methylated in (ENCODE) consortium provided genome-wide profiles of DNa- hepatocellular carcinoma. Ectopic overexpression of miR-1 seI hypersensitivity sites (DHS), which are additional surrogate reduced cell growth in vitro (16), which supported the function markers for regulatory elements such as promoters (23). As of miR-1-1 as a tumor suppressor. Although most of the studies data are available from a large panel of cell lines, the promoters focused on silencing of tumor suppressor miRNAs so far, of cell type–specific miRNA can be interrogated. Examples are activation of oncogenic miRNAs by epigenetic mechanisms the promoters of the muscle-specific miR-206 and the endo- has recently been shown to be equally important. Let-7a-3, for thelium-specific miRNA-126, which display DNaseI hypersen- example, was found methylated in normal lung tissue but not sitive site only in cell lines derived from muscle cells or methylated in lung adenocarcinoma. The hypomethylation endothelial tissue, respectively (24). correlated with high expression levels of let-7a-3 in lung The distance of miRNA promoter sequences to the miRNA cancer cell lines (17). A systematic overview of candidate coding sequence of the precursor miRNA can be up to 50 kb

OF2 Cancer Res; 73(2) January 15, 2013 Cancer Research

Epigenetic Regulation of miRNAs

(21); however, the likelihood of promoter presence increased A with increasing proximity (4). One third of the intragenic Hypermethylated miRNAs had an independent TSS. In cases where the distance 0.5 kb in CRC of the mature miRNA sequence to the host gene promoter was more than 10 kb, the likelihood for the presence of an inde- BC021736 pendent promoter increased, whereas miRNAs close to the BTG4 mir-34b–34c host gene promoter were frequently coexpressed with the host transcript (4). Like promoters of protein coding genes, miRNA CGI promoters are associated with TATA-box elements (19%), TFIIB recognition elements (21%) or an initiator (47%; ref. 4). B "MiRome"-wide promoter predictions suggested that more Hypermethylated Hypomethylated 5 kb than half of all miRNA TSSs are associated with a CpG island in solid tumors in CLL (55%–64%). In contrast, putative promoters of those miRNAs that overlap with previously annotated transcripts (e.g., host EF570049 mir-34a genes) have a CpG island frequency of 85% (4). CGI

Promoter DNA Methylation of Tumor C Suppressive and Oncogenic miRNAs 0.5 kb Hypomethylated The difficulty to characterize miRNA promoters and their in CLL epigenetic regulation is illustrated by members of the miR-34 family. These members are downstream targets of the p53 VMP1 mir-21 fi pathway (25) and were among the rst miRNAs for which Intron 10 Intron 11 aberrant promoter DNA methylation was characterized HeLa (6, 25, 26). Three family members are present in the mamma- lian genome: miR-34a is located on chromosome 1, miR-34b/c HEK293T reside on chromosome 11 as dicistronic cluster within transcription unit BC021736. Initially, BC021736 was considered Precursor miRNA (mir) Intron/exon the hosting gene of the miR-34b/c cluster. The BC021736 Promoter Methylated Unmethylated CpG promoter contains 2 p53-binding motives, which were suspected to account for the p53 dependent regulation of miR- Primary miRNA GpG island (CGI) 34b/c (25). However, only miR-34c resides in one of 2 BC021736 – exons, whereas miR-34b covers the exon intron junction and, © 2012 American Association for Cancer Research hence, is neither part of the fully processed BC021736 transcript nor the spliced intron (Fig. 1A). Moreover, only the expression of miR-34b/c but not of BC021736 was reactivated Figure 1. Aberrant promoter methylation of miR-34b/c, miR-34a, and miR- upon pharmacologic demethylation in the CRC cell line 21. The location of the miRNAs is indicated by red arrows; suggested HCT116. As an alternative to the BC021736 promoter, a CpG promoters and the emerging primary miRNA transcripts are displayed by island, 4.5 kb upstream of the miRNA-coding sequence and orange arrows. Aberrant cancer-specific methylation is shown by lollipops fi overlapping with the TSS of BTG4, was suggested as the (hypermethylation, lled; hypomethylation, empty). For details, see text. promoter for miR-34b/c. Although BTG4 is transcribed in the opposite direction of miR-34b/c, the upstream CpG island upstream of the miR-34a coding sequence (28). The suspected locus has promoter function in both directions (26). Hyper- promoter does not contain a known p53-binding motif and, methylation of the CpG island was found to be a common thus, leaves room for speculation of a p53-independent regu- feature of leukemias, lymphomas, and solid tumors (e.g., lation of miR-34a. gastric cancer; ref. 13). MiR-21 is oncogenic and is upregulated in a large panel of The third member of the miR-34 family, miR-34a, resides in proliferative malignancies (1). MiR-21 is processed from a 3 kb the second exon of a 33.8 kb transcript (EF570049), which primary transcript overlapping with the terminal genomic contains p53-binding sites and a CpG island in the promoter sequence of the protein-coding gene VMP1 (Fig. 1C). Yet, VMP1 region (Fig. 1B; refs. 6). Hypermethylation of this CpG island in is not the host gene, as the coding sequence of miR-21 is located a variety of solid tumors (breast-, lung-, colon-, kidney-, blad- downstream of VMP1. The promoter of miR-21 has been well der-, and pancreatic carcinoma) correlated with silenced studied in different cell systems, but different putative pro- expression (27). In CLL, miR-34a was upregulated, despite moters were defined depending on the analyzed tissue. In the same DNA methylation levels of the EF570049 promoter CpG cervical cancer cell line, HeLa, a region spanning from part of island in healthy B cells and CLL samples (28). Instead, loss of intron 10 into part of intron 11 of VMP1 displayed promoter DNA methylation in patients with CLL and enrichment of activity (29). A more upstream region only in the tenth intron H3K4me3, a marker for promoter presence, in cell lines derived revealed promoter activity in HEK293T cells, derived from from patients with CLL was observed in a region located 12 kb human embryonic kidney, or the melanoma cell line UACC62

www.aacrjournals.org Cancer Res; 73(2) January 15, 2013 OF3

Baer et al.

(4, 30). The apparently contradictory results indicate different DNA demethylation to assess epigenetic regulation of miRNA, tissue-specific miR-21 promoters. Tissue specificity of promo- the impact of aberrant hypomethylation in reactivation of ters should be taken into consideration when analyzing aber- miRNAs might have been underestimated. rant promoter methylation in disease. In our CLL study, the promoter sequence closer to miR-21, Outlook fi overlapping with the promoter identi ed in HeLa, was Large collaborations such as the International Cancer completely unmethylated in all CLL samples but showed Genome Consortium (ICGC) and The Cancer Genome Atlas fi signi cant DNA methylation in B-cell controls. This was in (TCGA) currently create extensive data sets of genetic, epige- line with upregulation of miR-21 in CLL cases compared with netic, and transcriptome profiles of different tumor entities their healthy counterparts. The fact that miR-21 is overex- and cell lines. Furthermore, the ENCODE consortium profiled a pressed in different tumor entities suggests DNA hypomethy- variety of cell lines for 12 histone modifications and variants lation as a major mechanism for miR-21 regulation. including H3K4me3 and acetylation of histone 3 at lysine 9 (H3K9ac) to disclose regulatory regions in the human genome. Epigenetic Regulation of miRNAs on a The resulting data will add to extend our knowledge on tissue- fi Genome-Wide Level speci c and ubiquitous miRNA promoters. As part of the ENCODE project, chromatin accessibility, a hallmark of pro- The number of known miRNAs has increased from around moters, and other regulatory elements, was defined by DHS in fi 400 in the rst promoter characterization in 2008 (20) to more 125 diverse cell and tissue types. Analysis of 329 miRNA than 1,400 in the recent miRBase 19 release (3). Thus, many of promoters revealed that 300 overlapped with or were close the recent genome-wide studies of miRNA regulation included to a DHS (24). The broad variety of tissues analyzed for DHS will extended screening for miRNA promoters. not only allow characterizing promoters of tissue-specific fi In mammary broblasts and epithelial cells, promoters for 232 miRNAs but might also permit to estimate the number of fi fi miRNAs were identi ed through H3K4me3 pro ling. Thirty-eight tissue-specific miRNAs promoters as suggested for miR-21. fi percent of these promoters showed tissue-speci c silencing by The importance of tissue specificity is underlined by results DNA methylation. For example, miR-200c/141 was only express- from parallel expression profiling, which revealed that only ed in epithelial cells and promoter methylation was evident in 28% of the known miRNAs were detectable in at least one of 15 fi broblasts. Other epigenetic-silencing mechanisms were found cell lines, and 59 miRNAs were expressed in one cell line only fi to be more frequent at the promoters of tissue speci cally (34). The ENCODE consortium is about to publish the genome- repressed miRNAs: 58% were silenced by histone 3 lysine 27 wide DNA methylation data completing analysis of epigenetic trimethylation, and redundancy of both repressive marks was regulation of all gene classes including miRNAs in cell lines. In fi limited to 21% of tissue speci cally silenced miRNAs (31). addition, cancer methylomes are analyzed and will be made fi H3K4me3 pro ling of the CRC cells line HCT116 and its publicly available by the ICGC, which, for example, provided derivative with a DNMT1 and DNMT3B double knockout the methylomes of patients with CLL (33). In conclusion, fi identi ed promoter data for 233 miRNAs. For 22 intragenic integrating data sets from different sources will enable scien- fi miRNAs, an independent, miRNA-speci c promoter was sug- tists to estimate the global influence of DNA methylation on fi gested and for the majority of intergenic miRNAs, the identi ed the regulation of miRNA and their aberrant behavior in cancer. promoter was within a 2 kb window. Comparison of the DNMT1 and DNMT3B double knockout and the wild-type cell Disclosure of Potential Conflicts of Interest line revealed epigenetic silencing of 37 miRNA promoters of No potential conflicts of interest were disclosed. which 22 contained a CpG island (32). DNA methylation was present at the promoter of miR-124 and miR-34b/c, which was Authors' Contributions previously identified to be silenced in HCT116 (12, 26) but also Conception and design: C. Baer, R. Claus, C. Plass for novel candidates, such as miR-338 or miR-944 (32). The Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): R. Claus function of both miRNAs and other recently identified silenced Writing, review, and/or revision of the manuscript: C. Baer, R. Claus, C. miRNAs (13) is poorly understood, but their epigenetically Plass Administrative, technical, or material support (i.e., reporting or orga- silencing could point towards a function as tumor suppressor. nizing data, constructing databases): R. Claus In our study, investigating epigenetic regulation of miRNAs fi in CLL, promoters for 781 miRNAs were identi ed by upstream Acknowledgments H3K4me3 enrichment using a panel of leukemic cell lines, The authors thank Dieter Weichenhan for critically reading the manuscript primary CLL cells, and healthy B cells (28). One hundred and and Yoon Jung Park for continued support and inspiring discussions. The authors apologize to those whose work was not included because of space twenty eight miRNAs were targets of recurrent aberrant meth- limitations. ylation in CLL cells compared with their healthy counterparts. Hypomethylation accounted for more than 60% of differential Grant Support methylation and was almost exclusively found adjacent to CpG The authors are supported by the German Carreras Foundation (DJCLS R 10/ islands or in CpG-free promoters. The high impact of hypo- 27). C. Baer holds a stipend of the Helmholtz International Graduate School for methylation is in concordance with the recent observations of Cancer Research. extensive genome-wide DNA hypomethylation in CLL (33). As Received September 23, 2012; revised November 19, 2012; accepted November previous studies used reactivation through pharmacologic 27, 2012; published OnlineFirst January 10, 2013.

OF4 Cancer Res; 73(2) January 15, 2013 Cancer Research

Epigenetic Regulation of miRNAs

References 1. Lujambio A, Lowe SW. The microcosmos of cancer. Nature 2012;482: 18. Lopez-Serra P, Esteller M. DNA methylation-associated silencing of 347–55. tumor-suppressor microRNAs in cancer. Oncogene 2012;31:1609–22. 2. Baylin SB, Jones PA. A decade of exploring the cancer epigenome - 19. Cheung HH, Davis AJ, Lee TL, Pang AL, Nagrani S, Rennert OM, et al. biological and translational implications. Nat Rev Cancer 2011;11: Methylation of an intronic region regulates miR-199a in testicular tumor 726–34. malignancy. Oncogene 2011;30:3404–15. 3. Kozomara A, Griffiths-Jones S. miRBase: integrating microRNA 20. Marson A, Levine SS, Cole MF, Frampton GM, Brambrink T, Johnstone annotation and deep-sequencing data. Nucleic Acids Res 2011;39: S, et al. Connecting microRNA genes to the core transcriptional D152–7. regulatory circuitry of embryonic stem cells. Cell 2008;134:521–33. 4. Ozsolak F, Poling LL, Wang Z, Liu H, Liu XS, Roeder RG, et al. 21. Chien CH, Sun YM, Chang WC, Chiang-Hsieh PY, Lee TY, Tsai WC, Chromatin structure analyses identify miRNA promoters. Genes Dev et al. Identifying transcriptional start sites of human microRNAs based 2008;22:3172–83. on high-throughput sequencing data. Nucleic Acids Res 2011;39: 5. Lee Y, Jeon K, Lee JT, Kim S, Kim VN. MicroRNA maturation: stepwise 9345–56. processing and subcellular localization. EMBO J 2002;21:4663–70. 22. Lee Y, Kim M, Han J, Yeom KH, Lee S, Baek SH, et al. MicroRNA genes 6. Tarasov V, Jung P, Verdoodt B, Lodygin D, Epanchintsev A, Menssen are transcribed by RNA polymerase II. EMBO J 2004;23:4051–60. A, et al. Differential regulation of microRNAs by p53 revealed by 23. Bernstein BE, Birney E, Dunham I, Green ED, Gunter C, Snyder M. An massively parallel sequencing: miR-34a is a p53 target that induces integrated encyclopedia of DNA elements in the human genome. apoptosis and G1-arrest. Cell Cycle 2007;6:1586–93. Nature 2012;489:57–74. 7. Hill DA, Ivanovich J, Priest JR, Gurnett CA, Dehner LP, Desruisseau D, 24. Thurman RE, Rynes E, Humbert R, Vierstra J, Maurano MT, Haugen E, et al. DICER1 mutations in familial pleuropulmonary blastoma. Science et al. The accessible chromatin landscape of the human genome. 2009;325:965. Nature 2012;489:75–82. 8. Heravi-Moussavi A, Anglesio MS, Cheng SW, Senz J, Yang W, Pren- 25. He L, He X, Lim LP, de Stanchina E, Xuan Z, Liang Y, et al. A microRNA tice L, et al. Recurrent somatic DICER1 mutations in nonepithelial component of the p53 tumour suppressor network. Nature 2007;447: ovarian cancers. N Engl J Med 2012;366:234–42. 1130–4. 9. Melo SA, Ropero S, Moutinho C, Aaltonen LA, Yamamoto H, Calin GA, 26. Toyota M, Suzuki H, Sasaki Y, Maruyama R, Imai K, Shinomura Y, et al. et al. A TARBP2 mutation in human cancer impairs microRNA proces- Epigenetic silencing of microRNA-34b/c and B-cell translocation gene sing and DICER1 function. Nat Genet 2009;41:365–70. 4 is associated with CpG island methylation in colorectal cancer. 10. Melo SA, Moutinho C, Ropero S, Calin GA, Rossi S, Spizzo R, et al. A Cancer Res 2008;68:4123–32. genetic defect in exportin-5 traps precursor microRNAs in the nucleus 27. Lodygin D, Tarasov V, Epanchintsev A, Berking C, Knyazeva T, Korner of cancer cells. Cancer Cell 2010;18:303–15. H, et al. Inactivation of miR-34a by aberrant CpG methylation in 11. Saito Y, Liang G, Egger G, Friedman JM, Chuang JC, Coetzee GA, et al. multiple types of cancer. Cell Cycle 2008;7:2591–600. Specific activation of microRNA-127 with downregulation of the proto- 28. Baer C, Claus R, Frenzel LP, Zucknick M, Park YJ, Gu L, et al. Extensive oncogene BCL6 by chromatin-modifying drugs in human cancer cells. promoter DNA hypermethylation and hypomethylation is associated Cancer Cell 2006;9:435–43. with aberrant microRNA expression in chronic lymphocytic leukemia. 12. Lujambio A, Ropero S, Ballestar E, Fraga MF, Cerrato C, Setien F, et al. Cancer Res 2012;72:3775–85. Genetic unmasking of an epigenetically silenced microRNA in human 29. Cai X, Hagedorn CH, Cullen BR. Human microRNAs are processed cancer cells. Cancer Res 2007;67:1424–9. from capped, polyadenylated transcripts that can also function as 13. Kunej T, Godnic I, Ferdin J, Horvat S, Dovc P, Calin GA. Epigenetic mRNAs. RNA 2004;10:1957–66. regulation of microRNAs in cancer: an integrated review of literature. 30. Fujita S, Ito T, Mizutani T, Minoguchi S, Yamamichi N, Sakurai K, et al. Mutat Res 2011;717:77–84. miR-21 Gene expression triggered by AP-1 is sustained through a 14. Vrba L, Jensen TJ, Garbe JC, Heimark RL, Cress AE, Dickinson S, et al. double-negative feedback mechanism. J Mol Biol 2008;378:492–504. Role for DNA methylation in the regulation of miR-200c and miR-141 31. Vrba L, Garbe JC, Stampfer MR, Futscher BW. Epigenetic regulation of expression in normal and cancer cells. PLoS One 2010;5:e8697. normal human mammary cell type-specific miRNAs. Genome Res 15. Davalos V, Moutinho C, Villanueva A, Boque R, Silva P, Carneiro F, et al. 2011;21:2026–37. Dynamic epigenetic regulation of the microRNA-200 family mediates 32. Suzuki H, Takatsuka S, Akashi H, Yamamoto E, Nojima M, Maruyama epithelial and mesenchymal transitions in human tumorigenesis. R, et al. Genome-wide profiling of chromatin signatures reveals epi- Oncogene 2012;31:2062–74. genetic regulation of MicroRNA genes in colorectal cancer. Cancer 16. Datta J, Kutay H, Nasser MW, Nuovo GJ, Wang B, Majumder S, et al. Res 2011;71:5646–58. Methylation mediated silencing of MicroRNA-1 gene and its role in 33. Kulis M, Heath S, Bibikova M, Queiros AC, Navarro A, Clot G, et al. hepatocellular carcinogenesis. Cancer Res 2008;68:5049–58. Epigenomic analysis detects widespread gene-body DNA hypo- 17. Brueckner B, Stresemann C, Kuner R, Mund C, Musch T, Meister M, methylation in chronic lymphocytic leukemia. Nat Genet 2012;44: et al. The human let-7a-3 locus contains an epigenetically regulated 1236–42. microRNA gene with oncogenic function. Cancer Res 2007;67: 34. Djebali S, Davis CA, Merkel A, Dobin A, Lassmann T, Mortazavi A, et al. 1419–23. Landscape of transcription in human cells. Nature 2012;489:101–8.

www.aacrjournals.org Cancer Res; 73(2) January 15, 2013 OF5

Genome-Wide Epigenetic Regulation of miRNAs in Cancer

Constance Baer, Rainer Claus and Christoph Plass

Cancer Res Published OnlineFirst January 10, 2013.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-12-3731

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2013/01/10/0008-5472.CAN-12-3731. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.